AI-Native Developer

Role Description

You will join an existing team and take technical ownership of a portfolio of live products and SaaS, while spending most of your energy building new things that did not exist before. The balance we aim for is roughly seventy percent innovation — new agents, new integrations, new capabilities — and thirty percent maintaining and evolving what is already running. That split is a principle, not a rigid rule: when a live system needs attention, you give it. What stays constant is that you own your output end to end, whether it is new or existing.

The binding constraint on what you produce is no longer typing speed. It is judgment — your ability to direct AI tools toward correct, production-ready output, validate what comes back, and iterate with precision.

What You Will Own

• Features and integrations that ship to production — real users, real consequences, and your name on them
• Quality of AI-generated code before it merges — you develop the instinct to distinguish correct from plausible, and you apply it before anything reaches production
• LLM integrations that hold under real conditions — connecting Claude, GPT, or other models via OutSystems' secure API gateway in a way that works in production, not just in demos
• AI agents built on OutSystems Agent Workbench that run unsupervised on real enterprise workflows and produce output that is trustworthy, not just impressive
• Your share of the existing portfolio — understanding the applications the team maintains well enough to extend them without accumulating hidden debt
• Shared context that makes the whole team faster — documenting decisions and patterns so that the team's AI tools operate with full enterprise context; treating documentation as compound interest

The Behaviors We Are Looking For

• You build first and sharpen later.
• You distinguish correct from plausible.
• You own what you ship.
• You find a way around walls.
• You share context generously.
• You are honest about what you do not know.

Experience

• 3+ years of development experience, with depth in at least one area — integration, data, UI — and genuine curiosity across others
• Active practitioner of AI-augmented development — you use Claude, Cursor, Copilot, or similar tools as a core part of how you build, not as an occasional shortcut
• Strong integration fundamentals: REST APIs, JSON, authentication patterns, error handling under real conditions
• Comfortable producing something from a rough brief and sharpening it through iteration

Bonus

• Experience with OutSystems Mentor or AI Agent Builder
• Familiarity with prompt engineering patterns and LLM context window management
• Prior experience in a high-autonomy environment where every person's output is visible and tested daily
• Domain knowledge in finance, insurance, healthcare, or logistics

This Is Not For You If...

• You treat AI tools as an occasional shortcut rather than the primary way you build
• Ambiguity stalls you — you need complete requirements before you can start
• You consider maintaining and evolving existing systems to be below your level
• You have never caught an AI tool producing output that looked right but wasn't — or you have, and it did not change how carefully you review
• You prefer to discuss what to build over shipping something testable and learning from it

Growth Path

High performers in this role move into technical leadership within 12 to 18 months, or into specialist roles in AI agent design or enterprise context engineering — disciplines that barely existed two years ago and are among the highest-leverage skills in the industry today. We will show you who has already made that move.

Why Join

The skills you will develop here — context engineering, AI agent validation, enterprise-grade agentic development — are not in any curriculum. You will learn them by applying them in production, on real systems, alongside people who are figuring them out at the same frontier.

If you want to be one of the first people in the industry who actually knows how this works — not in theory, but in production — this is where you do it.